期刊文献+

一种支持向量机的组合核函数 被引量:22

Compound kernel function of support vector machines
下载PDF
导出
摘要 核函数是支持向量机的核心,不同的核函数将产生不同的分类效果,核函数也是支持向量机理论中比较难理解的一部分。通过引入核函数,支持向量机可以很容易地实现非线性算法。首先探讨了核函数的本质,说明了核函数与所映射空间之间的关系,进一步给出了核函数的构成定理和构成方法,说明了核函数分为局部核函数与全局核函数两大类,并指出了两者的区别和各自的优势。最后,提出了一个新的核函数———组合核函数,并将该核函数应用于支持向量机中,并进行了人脸识别实验,实验结果也验证了该核函数的有效性。 Kernel functions are key part and the hard issue of Support Vector Machines (SVM). Different kernel can produce different SVM. SVM can deal with nonlinear problems in classification and Regression easily by using kernel functions. In the paper, we discussed the relationship between kernel functions and nonlinear mappings and mapped spaces. Then we introduced a rule of constructing kernel function. At last, we proposed a new kernel -- compound kernel function. We applied the kernel in SVM and compared the kernel with other kernels in human face recognition experiment. Experimental results certify that the new kernel can achieve better performance than other kernels.
作者 张冰 孔锐
出处 《计算机应用》 CSCD 北大核心 2007年第1期44-46,共3页 journal of Computer Applications
基金 暨南大学引进人才基金资助项目(04JZKY005)
关键词 支持向量机 核函数 组合核函数 基于核的学习 SVM( Support Vector Machines) kernel function compound kernel function kernel-based learning algorithms
  • 引文网络
  • 相关文献

参考文献9

  • 1VAPNIK VN.The Nature of Statistical Learning Theory[M].New York:Springer Verlag,1995.
  • 2BURGES CJC.A tutorial on support vector machines for pattern recognition[J].Knowledge Discovery and Data Mining,1998,2(2):121-167.
  • 3MULLER KR,MIKA S,G.RTSCH,et al.An Introduction to Kernel-Based Learning Algorithms[J].IEEE Transactions on Neural Networks,2001,12(2):181-201.
  • 4SCHOLKOPF B,MIKA S,BURGES CJC,et al.Input space versus feature space in kernel-based methods[J].IEEE Transactions on Neural Networks,1999,10(5):1000-1017.
  • 5SCHOLKOPF B,SMOLA AJ.Learning with Kernels[M].MA:MIT Press,2001.
  • 6SAUNDERS C,STITSON MO,WESTON J,et al.Support vector machine reference manual[R].Technical Report CSD-TR-98-03.Royal Holloway University,1998.
  • 7SMITS GF,JORDAN EM.Improved SVM regression using mixtures of kernels[A].IEEE Proceeding of IJCNN '02 on Neural Networks[C].2002,3.2785-2790.
  • 8SCHOLKOPF B.The Kernel Trick for Distances[R].Technical Report MSR-TR-2000-51,Microsoft Research,2000.
  • 9CRISTIANINI N,SHAWE-TAYLOR J.An Introduction to Support Vector Machines and Other Kernel-based Learning Methods[M].The Syndicate of the Press of the University of Cambridge,2000.

同被引文献173

引证文献22

二级引证文献159

;
使用帮助 返回顶部